Northwestern Polytechnical University, a Chinese university on the US Entity List since 2001, bought two Nvidia H200 chips in October for 874,700 yuan ($123,000), Reuters reported on 11 November, citing an October 21 procurement record on the university’s website. That is 437,350 yuan per chip, by our arithmetic.
The chips were bought for the university’s management school from Shaanxi Bosuan Network Technology rather than directly from Nvidia, according to Reuters. Northwestern Polytechnical, Nvidia and the Commerce Department’s Bureau of Industry and Security did not respond to Reuters’ requests for comment.
Business Pill · GPUs Explained
An evergreen Business Engineer explainer on why AI runs on GPUs, the export-controlled chips in these records. It teaches the general idea only and does not discuss the Reuters report.
The key insight: As we read it, the records describe a buyer that combined two routes to the same chip: Northwestern Polytechnical bought two H200 chips outright for 874,700 yuan and paid 882,700 yuan to rent access to 16 more, by our arithmetic on the figures Reuters reports.
Reuters says it reviewed 26 Chinese procurement and corporate records from this year, which show universities and research institutions buying US-restricted Nvidia chips, or renting access to computing services that use them, through domestic suppliers. Several had close links to China’s defence industry or military research, according to the report.
At Northwestern Polytechnical, the cybersecurity school awarded two seven-month contracts in January for access to servers containing eight H200 chips each, worth 441,700 yuan and 441,000 yuan, the records show, according to Reuters. The university also rented access to servers containing eight Nvidia A100 chips and eight A800 chips.
Elsewhere, Shenzhen University bought three H100 processors for 769,800 yuan in September, and Beihang University bought four A800 processors for 489,000 yuan in April, Reuters reported. Wuhan University of Technology bought a server containing an H100 chip for 355,000 yuan in June and four A100 chips for 344,000 yuan.
Reuters says the awards went to several little-known Chinese technology suppliers, and that the records offered scant information about where the processors originated or where the rented servers operated.
The nine 2026 procurement records with stated yuan values in Reuters’ 11 November report. Northwestern Polytechnical University (two H200 chips, 874,700 yuan; two 8-H200 server-access contracts, 441,700 and 441,000 yuan; an April contract for computing needing at least 24 A100 chips, 375,000 yuan) and Beihang University (four A800 chips, 489,000 yuan), in red, appear on the US Entity List. The 4,575,000-yuan total and the 57.3% share are our arithmetic.
The Entity List Entry
The current Entity List text in the eCFR, which we read, lists Northwestern Polytechnical University with ten aliases and a licence review policy of “Presumption of denial”. Its first citation is 66 FR 24266 of 14 May 2001, and the entry carries eight Federal Register citations, the latest dated 16 September 2025.
Beihang University appears in the same list as an alias of Beijing University of Aeronautics and Astronautics, whose entry also starts with 66 FR 24266 of 14 May 2001. Reuters says both universities belong to the “Seven Sons of National Defense” group.
According to Reuters, the Trump administration began allowing some H200 exports to China case by case in January, but that policy did not lift the separate restrictions on Entity List organisations. The January rules also require safeguards against blacklisted entities and military end-users remotely accessing chips exported under the policy, the report says.
Northwestern Polytechnical University’s three valued H200 records in 2026, as reported by Reuters: two chips bought outright for 874,700 yuan and two seven-month contracts for access to servers with eight H200 chips each, for 441,700 and 441,000 yuan. Shares and the per-chip and per-month figures are our arithmetic.
Our Numbers
The nine records in the Reuters report that carry a stated yuan value total 4,575,000 yuan, by our arithmetic. Of that, 2,621,400 yuan, or 57.3%, went to the two buyers we found on the Entity List: Northwestern Polytechnical’s four valued records, counting the April contract for computing that needs at least 24 A100 chips, and Beihang’s A800 purchase.
Northwestern Polytechnical’s two rental contracts cover 16 H200 chips for seven months each, or 112 chip-months, for 882,700 yuan. That is about 7,881 yuan per H200 per month, by our arithmetic, so the 437,350 yuan paid per chip bought outright equals about 55 months of rented access.
The H200 price per chip is about 1.7 times the 256,600 yuan per chip implied by Shenzhen University’s three-H100 purchase, by the same arithmetic. Both figures come from single procurement records, not market prices.
The Structural Read
As we read the Entity List entry, the October purchase reached a university whose listing dates to 14 May 2001 and whose licence policy is a presumption of denial, the setting in which Reuters says licences are normally denied.
As we read the price record, renting looked cheap against owning: about 7,881 yuan per H200 per month against 437,350 yuan per chip bought, or about 55 months of rented access per chip owned, by our arithmetic.
Reuters, 11 November 2026
“Under Chinese law, public procurement agencies must verify that suppliers can fulfil orders before announcing winning bidders and signing contracts.”
Three Implications
FOR EXPORT CONTROLS Reuters reports that the January policy allowing some H200 exports to China case by case did not lift the separate restrictions on Entity List organisations such as Northwestern Polytechnical.
FOR NVIDIA The October chips were bought from Shaanxi Bosuan Network Technology rather than directly from Nvidia, the record shows, according to Reuters; Nvidia did not respond to Reuters’ request for comment.
FOR CONGRESS Reuters says Democrats and Republicans have urged stronger controls on China’s access to advanced AI chips, such as chip-tracking requirements and greater congressional oversight of exports.
The framework argues that the AI supply chain is “creating geopolitical leverage that operates independently of who has the best models”.
As we read it, an Entity List entry is one tool of that leverage, and these records show how far it reaches when buyers use domestic resellers and rented servers. The framework predates the Reuters report and does not discuss it.
What Is Not Established
The records, as Reuters describes them, do not show where the chips originated or how they reached the Chinese suppliers, and Reuters could not determine contact details for Shaanxi Bosuan Network Technology. We did not open the procurement records ourselves; every record figure here is as Reuters reports it.
The rental contracts covered fine-tuning AI models and reinforcement-learning training, so their price may include services beyond chip access; our per-month figure divides the contract value by chip-months only. In the eCFR text we searched, Shenzhen University, Wuhan University of Technology and the University of Chinese Academy of Sciences did not match under those names.
This Week In The AI Supercycle
The Great Convergence of Capital, Power, and Geopolitics
This week’s issue: for two years the AI build-out was priced as a supply story, and last week as a credit story. This week it was priced as a permission story, and the answer arrived as a cost, not a ban.
A university on the US Entity List since 2001 bought two Nvidia H200 chips in October for 874,700 yuan, or 437,350 yuan per chip, according to procurement records reported by Reuters. Across the nine valued records in the report, 57.3% of the yuan went to the two Entity-listed universities, by our arithmetic.
A note on sourcing. We read the Reuters report of 11 November 2026 in full through a syndicated copy (the Fulton Sun), because reuters.com returned a captcha, and the Entity List text in the eCFR, current to 9 November 2026. We did not open the Chinese procurement records; record figures are as Reuters reports them. Per-chip, per-month and share figures are our arithmetic. Reuters says Northwestern Polytechnical, Nvidia and the Bureau of Industry and Security did not respond to its requests for comment; we did not contact them for this piece.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
